Publication Search

This study aims to develop an Indonesian traffic sign detection system using a transfer learning approach to improve road safety and traffic efficiency. The dataset was obtained from Kaggle and consists of 2,100 images across 21 traffic sign classes. The research stages include data collection, preprocessing to reduce noise and normalize image brightness, object detection using YOLOv5, and classification based on transfer learning with ResNet, VGG-16, and MobileNet architectures. Model performance was evaluated using accuracy, precision, recall, and F1-score metrics. Experimental results indicate that the YOLOv5 model is capable of detecting traffic sign objects; however, the classification performance remains relatively low, with a mean Average Precision (mAP) value of 0.17. These findings suggest that further optimization is required in data preprocessing, dataset quality, and model parameter tuning to achieve better performance. This study demonstrates that transfer learning has significant potential for developing computer vision-based traffic sign detection systems, although further improvements are necessary to ensure robustness under real-world Indonesian traffic conditions.

Program strategis nasional (PSN) in Indonesia public welfare through social, economic,environmental, and governance dimensions. Findings indicate that limiting truck operations during peak hours generally contributes to reduced traffic congestion, improved road safety, and lower levels of pollution and noise in residential areas. However, the effectiveness is constrained by weak enforcement, frequent rule violations by operators, and limited crosssector coordination. To enhance the policy's positive impact, this study recommends strengthening law enforcement, utilizing monitoring technologies, and developing a more integrated and participatory logistics system. When consistently and collaboratively implemented, the policy has significant potential to support the success of NSPs and improve community welfare.

This study analyzes noise levels caused by traffic activities and generator sets (genset) in the Semarang University environment. Measurements were conducted using descriptive analytical methods by comparing measurement results against noise quality standards according to KepmenLH No.48 of 1996. Data collection was carried out in November 2023 at three location points (in front of Building V, Sports Center, and USM gate) during morning (08.00-08.30), afternoon (12.30-13.00), and evening (17.30-18.00) on weekdays and weekends. Measurement results showed traffic noise levels on weekdays reached 88.3 dB in front of the Sports Center, 85.3 dB at Building V, and 82.1 dB at the USM gate. On weekends, the highest noise level reached 82 dB. Generator noise measurements in stationary conditions reached 108.4 dB and increased to 112.4 dB during loading. The highest vehicle volume occurred during weekday evenings, reaching 3,176 vehicles/hour. Research results indicate that noise levels in the campus environment are above the permitted threshold of 55 dB for educational areas. The study recommends the construction of noise barriers and evaluation of generator placement to reduce noise impact on campus activities.

Noise in educational environments can affect the effectiveness of teaching and learning processes. This study aims to analyze the impact of traffic noise on teaching and learning activities in Building V of Semarang University, located along the Soekarno-Hatta arterial road. The research method uses a qualitative descriptive approach by measuring noise levels using a Sound Level Meter on six floors of the building. Measurements were taken during active lecture hours with time intervals of 3 x 30 minutes. The results showed that noise levels on floors 1 to 3 were above the educational environment quality standard of 55 dB according to KepMenLH No. 48 of 1996. Meanwhile, on the 6th floor, the noise level averaged around 55 dB. Excessive noise makes it difficult for students to hear lecturers' explanations, reduces their ability to capture information, and decreases learning material retention. To address these issues, sound insulation installation, spatial planning improvements by designing room layouts farther from noise sources, vegetation planting, and lecture rescheduling to avoid peak traffic times are necessary.

The detection of computer network attacks is becoming increasingly important as the complexity and frequency of cyber-attacks threatening information systems and network infrastructure continue to rise. These attacks may lead to severe consequences, including data breaches, service disruptions, and financial losses. To address these challenges, artificial intelligence techniques have become a major focus in the development of more effective, adaptive, and reliable intrusion detection systems. Among various classification algorithms, the C4.5 decision tree has demonstrated strong performance due to its simplicity, interpretability, and high classification accuracy. This study aims to apply the C4.5 algorithm for network attack detection using a comprehensive dataset that includes multiple categories of attacks and normal network activities. The proposed methodology consists of several stages, including data preprocessing, feature selection, decision tree model construction, and performance evaluation using standard metrics such as accuracy, precision, recall, and F1-score. Data preprocessing is performed to handle missing values, normalize data, and reduce noise, thereby improving the overall quality of the dataset and enhancing classification results. The experimental results indicate that the C4.5 decision tree algorithm effectively classifies network traffic into attack and normal categories with a satisfactory level of accuracy. The model successfully identifies attack-related patterns and highlights significant features that influence detection performance. Further analysis reveals that appropriate feature selection and parameter tuning significantly contribute to improving model reliability and robustness. This research provides a valuable contribution to the development of efficient, accurate, and practical network intrusion detection systems. The proposed approach is expected to strengthen information security frameworks and support proactive defense strategies against increasingly sophisticated cyber threats, thereby enhancing the protection of critical network infrastructures.

The purpose of this service activity is to provide an increased understanding of the concept of noise and carbon emissions of motorized vehicles located on Golf Course Road. Traffic noise and carbon emissions are environmental problems that have an impact on public health. The method used is field observation by measuring the number of vehicles, types of vehicles, noise levels using a decibel meter, and estimating carbon emissions using an estimative approach based on the number of vehicles and the average carbon emissions of each type of vehicle. This service shows the importance of periodic monitoring of noise and air pollution as a basis for urban environmental policy making. This activity went well, so that it was able to provide an increase in the number of vehicles.

There are 514 urban areas in the archipelagic nation of Indonesia (Ministry of Home Affairs, 2016). There are many cities in Indonesia that are able to provide a sense of comfort and security for their residents, this of course has an impact on population growth. However, the amount of waste that can be generated and stored is limited by the carrying capacity of the urban environment. Using the idea of ​​a "green city" which consists of nine elements of road infrastructure, traffic, green open space, clean water, noise, energy, housing, clean air and buildings is one way to overcome the above problems. All these qualities need to be utilized immediately with a comprehensive systems approach and in accordance with economic, social and environmental principles in sustainable development. Blue economy approaches can be used to eliminate waste. The “blue economy” is a term that is becoming increasingly popular in today's seas and oceans. The aim of this concept is to combine sea base development opportunities with environmental management and protection. Four conceptual interpretations of the blue economy were determined through research into the dominant discourse in international economic policy documents. How the blue economy works is also examined through an “in practice” analysis of the blue economy and related actors. Next, the scope and focus of the blue economy is explored specifically on the maritime industry, which is included or excluded in various conceptualizations. This analysis reveals areas of agreement and conflict. The consensus range reflects the growing trend towards the commercialization and valuation of nature, delineating and defining maritime boundaries and enhancing the security of the world's oceans. There are several areas of conflict, particularly disputes over the legitimacy of individuals as part of the “blue economy”, highly carbon-intensive industries such as oil and gas and the growing deep sea mining industry. Oceans are becoming increasingly important in terms of potential international trade opportunities through intermediaries and buyers. the behavior of the model supporting such a relationship in the ocean is acceptable from a regular and economic point of view. The relationship between land and sea is increasing in the role and importance of the "blue economy" as the term emerges in the background. . A literature review was conducted to analyze the correct definition of the blue economy. This definition is analyzed based on the minimum requirements that are important for the blue economy. This article also tries to compile various types of activities related to marine services. This is done to determine what can be considered priority areas for blue economic growth. The blue economy concept is based on national economic development. This is comprehensively useful for achieving overall national development. Environmental accounting is stated as a process of reducing negative values ​​and creating positive values ​​in environmental accountability.